Accelerometry Annotation Tool

A web-based application for visualizing and annotating accelerometry data, built with Panel and Bokeh. The app is sensor-agnostic — it works with any tri-axial accelerometry data stored in HDF5 format. Developed by the NSHAP Lab at the University of Chicago.

Key features

  • LTTB downsampling — renders 500K+ data points per axis as ~10,000 visually representative points using the Largest Triangle Three Buckets algorithm

  • Box-select annotation — click and drag on the signal plot to annotate activity episodes with configurable labels

  • Segment / Scoring / Review flags — mark individual repetitions, select the best segment for scoring, and flag ambiguous signals for peer review

  • Multi-user collaboration — deterministic file assignment, per-user annotation files, admin impersonation, and role management

  • Auto-save to Excel — one-click export to per-user Excel files in data/output/

  • HPC deployment — self-service connect script for Slurm clusters with automatic SSH tunneling

Quick start

panel serve visualize_accelerometry/app.py \
    --port 5601 \
    --basic-auth credentials.json \
    --cookie-secret $(python -c "import secrets; print(secrets.token_hex(32))") \
    --allow-websocket-origin localhost:5601 \
    --basic-login-template visualize_accelerometry/templates/login.html

Then open http://localhost:5601/app in your browser.